Linux Foundations

Every Linux Foundations lesson in the Linux for AI Engineering course, in the order you study them. Each page opens about the first half of its lesson to read, some the whole of it; the full lessons are in the plan.

8 lessons

  1. Linux for AI: the GPU driver path, distribution choice and lab bring-up

    Why an NVIDIA driver installs but finds no GPU in a virtual machine, why AI runs on Linux, and how the GPU reaches your code.

  2. Headless home AI servers: power draw, Wake-on-LAN and models on a USB stick

    Turning an old PC into a headless home AI server: what it costs to leave on, sleep and Wake-on-LAN, and running models from a USB stick.

  3. The shell, the FHS and POSIX permissions: groups, setgid and shared model directories

    How Linux decides who may read a file, why a new group needs a new login, hard and soft links, and a folder a team can share.

  4. Streams and pipelines: stdout and stderr, grep, jq and batch processing at scale

    Why a program prints nothing until it is piped somewhere else, stdout and stderr, redirection order, and building shell pipelines.

  5. LLMs in the shell: piping context to a model and guarding the commands it writes

    Using a language model as one more tool in a Linux pipe: context windows and tokens, whether input will fit, and scripts you can trust.

  6. Bash scripting: exit codes, strict mode, safe checkpoint pruning and cron

    Why a bash script carries on after a failed command, exit codes, set -euo pipefail, pruning checkpoints safely and running on a schedule.

  7. Processes and signals: graceful shutdown, keeping jobs alive, limits and the OOM killer

    Why GPU memory stays in use after kill -9, process trees and orphans, SIGTERM versus SIGKILL, and stopping training jobs gracefully.

  8. Packages and shared libraries: dynamic linking, version pinning, source builds and GLIBCXX errors

    Why a copied program fails with GLIBC version not found, how the dynamic linker finds libraries, package managers and pinning versions.